NDA 211723: Tazemetostat

FDA Opening Remarks Ashley Ward, MD Division of 3 Office of Oncologic Diseases December 18, 2019 Proposed Indication

Tazemetostat is indicated for the treatment of patients with metastatic or locally advanced epithelioid who are not eligible for curative surgery

www.fda.gov 2 • Rare soft tissue sarcoma • Approximately 125 cases per year in the U.S. • Patients typically diagnosed between 20-40 years of age • Approximately 50% have metastatic disease at diagnosis • 5-year survival of patients with metastatic disease: 0% • Approximately 90% show nuclear loss of integrase interactor 1 (INI-1)

www.fda.gov 3 Therapeutic Approaches • Wide surgical excision, with or without radiation • Systemic chemotherapy reserved for metastatic disease • FDA-approved therapies for soft-tissue sarcoma: – Doxorubicin (1974) – Pazopanib (2012)

www.fda.gov 4 Tazemetostat • First-in-class, oral, small molecule inhibitor of the methyltransferase enhancer of zeste homolog 2 (EZH2) • Postulated to restore balance to a set of proteins involved in chromatin remodeling and gene expression in tumors that have lost the integrase interactor 1 (INI-1) • Impact of this on cell biology not well understood

www.fda.gov 5 Study EZH-202: Design

Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5 Cohort 6 Cohort 7 Rhabdoid Synovial INI-1 Loss Renal Epithelioid Epithelioid Chordoma Tumors Sarcoma or EZH2 Medullary Sarcoma Sarcoma mutation

N=15 N=15 N=15 N=15 N=15 N=40 N=15 Stage 1 Stage

N=30 N=30 N=30 N=30 N=30 N=30 Stage 2 Stage

N=30 Expansion All patients received tazemetostat 800 mg by mouth twice daily www.fda.gov 6 Study EZH-202: Efficacy Results

Cohort 5 Cohort 6 Pooled N=62 N=44 N=106 ORR 15% 11% 13% (95% CI) (7, 26) (4, 25) (7, 21) CR (n, %) 1 (1.6) 1 (2) 2 (2) PR (n, %) 8 (13) 4 (9) 12 (11) Range of DOR 4, 24+ 3.5, 18.2+ 3.5, 24+ Median follow-up in months (range) 13.8 (0.2, 32) 11.8 (0.2, 21) 12.8 (0.2, 32)

1 By independent review according to RECIST v1.1 Key: ORR, Overall Response Rate; CI, Confidence Interval; CR, Complete Response; PR, Partial Response; DOR, Duration of Response www.fda.gov 7 Study EZH-202: Safety Results

Cohort 5 N=62 All grade adverse events 62 (100%) Grade 3-4 adverse events 30 (48%) Serious adverse events 23 (37%) Treatment discontinuations due to adverse events 1 (1.6%) Dose interruptions due to adverse events 21 (34%) Dose reductions due to adverse events 1 (1.6%)

Warning: Risk of Secondary Malignancies www.fda.gov 8 Issues

• Discuss whether the evidence from Cohorts 5 and 6 of EZH-202 is sufficient to establish the benefit of tazemetostat in patients with epithelioid sarcoma. – Observed ORR 11-15% (lower bound of 95% CI: 4-7%) • Does the demonstrated benefit of tazemetostat outweigh the risks of the drug in the proposed indication? – Uncertainty around risk profile due to single arm study design; 48% with Grade 3-4 adverse event, 37% with serious adverse event – Identified risk: secondary malignancies www.fda.gov 9

NDA 211723: TAZEMETOSTAT

Oncologic Drugs Advisory Committee Meeting Efficacy and Safety Analyses and Issues December 18, 2019

Leslie Doros, MD Clinical Reviewer Division of Oncology 3 Office of Oncologic Diseases www.fda.gov 11 Proposed Indication

Proposed Indication for Accelerated Approval: Tazemetostat is indicated for the treatment of patients with metastatic or locally advanced epithelioid sarcoma who are not eligible for curative surgery

Proposed Dosing Regimen: 800 mg orally twice daily with or without food

12 Key Issue for Consideration

Discuss whether the evidence from Study EZH-202 is sufficient to establish the benefit of tazemetostat in patients with epithelioid sarcoma.

13 Study Scheme for EZH-202 Cohorts 5 and 6

Original Cohort

Expansion Cohort

14 Source: Copied from Protocol EZH-202 Amendment 5 page 41 Rationale for Pooling Cohorts 5 and 6

• Eligibility criteria differences not expected to have large efficacy impact – Integrase interactor 1 (INI1) loss required for Cohort 5 – Tumor biopsy required for Cohort 6 – Must have progressed within 6 months of study entry for Cohort 6 • This criterion only applied to expansion stage of Cohort 5

• Similarities between Cohorts 5 and 6 – Demographic and baseline characteristics were similar • 34/44 patients in Cohort 6 had INI1 loss; 4 retained INI1 and 6 had missing INI1 status – Same dose and regimen of tazemetostat – Similar follow-up (using updated cut-off for Cohort 6) – Overall response rate (ORR) and duration of response (DOR) collected on both cohorts in same manner

15 Study EZH-202: Overall Response Rate

EZH-202 EZH-202 EZH-202 Cohort 5 Cohort 6 Pooled N=62 N=44 N=106 ORR1 15% 11% 13% (95% CI) (7, 26) (4, 25) (7, 21) CR (n, %) 1 (1.6) 1 (2) 2 (2) PR (n, %) 8 (13) 4 (9) 12 (11) 1As assessed by Response Evaluation Criteria for Solid Tumors (RECIST) v1.1 by independent review committee (IRC) Key: ORR, Overall Response Rate; CI, Confidence Interval; CR, Complete Response; PR, Partial Response 16 Study EZH-202: Duration of Response

Cohort 5 Cohort 6 Pooled N=62 N=44 N=106 Number of Responders 9 5 141 ≥ 6 months 6 3 9 ≥ 12 months 3 1 4 Range of DOR 4, 24+ 3.5, 18.2+ 3.5, 24+ Median follow-up in months (range) 13.8 (0.2, 32) 11.8 (0.2, 21) 12.8 (0.2, 32) 1Seven patients had an ongoing response at the time of the data cutoff. Key: DOR, Duration of Response 17 ORR per IRC by Subgroup

Subgroup # Responses N ORR1 (95% CI) Prior lines of therapy 0 7 44 16% (7%, 30%) 1+ 7 62 11% (5%, 22%) INI1 status at baseline Missing 0 6 0% (0%, 46%) Deficient 14 96 15% (8%, 23%) Present 0 4 0% (0%, 60%) Cohort Cohort 5 9 62 15% (7%, 26%) Cohort 5: Original cohort 6 31 19% (7%, 37%) Cohort 5: Expansion cohort 3 31 10% (2%, 26%) Cohort 6 5 44 11% (4%, 25%) 1ORR based on pooled data from Cohorts 5 and 6 except as indicated. Key: ORR, Overall Response Rate 18 Factors FDA Considers When Interpreting ORR

• Magnitude and durability of effect • Benefits and risks of other therapies used to treat the disease • Location, size, and clinical impact of tumor • Mechanism of drug action and underlying disease biology • Body of knowledge regarding the drug’s effects in other settings • Safety of the drug

19 FDA-Approved Drugs for Soft Tissue Sarcoma

Doxorubicin • Approved in 1974 for treatment in soft tissue sarcoma (STS) – Approval based on response rate1 of 24% (95% CI: [19, 30]) – Range of response rate in published studies is 8-19%

Pazopanib • Approved in 2012 for the treatment of STS after chemotherapy – Based on progression-free survival (PFS)2: • Hazard ratio: 0.35 (95% CI: [0.26, 0.48]) • Median PFS 4.6 months vs. 1.6 months in placebo arm – ORR: 4% (vs. 0%) – OS: 12.6 months (vs. 10.7 months)

1Response defined as a decrease of 50% in tumor mass. 2 https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/022465s-010S-012lbl.pdf 20 Activity in Patients with Epithelioid Sarcoma

Number of ES Patients Response Rate (95% CI) Pink, et al 20141 Anthracycline +/- Ifosfamide 13 0 (0, 25) Jones, et al 20122 Anthracycline + Ifosfamide 9 11 (0, 48) Anthracycline 10 20 (3, 56) Touati, et al 20183 Doxorubicin (1st line) 5 0 (0, 52) Doxorubicin + Ifosfamide (1st line) 8 13 (0, 53) Pazopanib (1st and 2nd line) 11 27 (6, 61) Frezza, et al (2018)3 Pazopanib 18 0 (0, 19) Anthracycline-based 85 22 (14, 33) 1 Response assessed by World health Organization (WHO) criteria and RECIST crit 2 Response assessed by RECIST criteria 3 Response assessed by RECIST 1.1 21 K ES ith li d Tumor Burden

Study EZH-202 did not collect data regarding clinical impact (e.g., cosmetic, functional) of baseline tumor burden • Most patients (74%) had 5 or fewer measurable tumors at baseline; maximum was 10 • 84% of non-nodal target lesions were 5 centimeter (cm) or smaller in longest diameter; only 3 non- nodal target lesions were >10 cm

22 Mechanism of Action and Tumor Response • The methyltransferase enhancer of zeste homolog 2 (EZH2) catalyzes the activity of histone H3, generally downregulating transcription • INI1 loss can lead to abnormal activity/expression of EZH2 and an oncogenic dependence on EZH2 • Tazemetostat inhibits EZH2, restoring transcriptional homeostasis • Low response rate to tazemetostat in patients with INI1 deficient epithelioid sarcoma could be due to: – Lack of relevance of the target in this disease, or – Effects that inhibit tumor cell growth rather than cause tumor cell death • Activity of tazemetostat in : – ORR of 69% for patients with EZH2 mutation; 35% for patients with wild-type EZH2 23 Safety Summary for Cohort 5

• Most common treatment-emergent adverse events (TEAEs): – Pain, fatigue, nausea, decreased appetite, vomiting, and constipation • Grade 3 to 4 TEAEs: 30 (48%) patients – Anemia, pain, decreased weight, decreased appetite, hemorrhage, dyspnea, and pleural effusion • Serious AEs: 23 (37%) patients – Hemorrhage, pleural effusion, dyspnea, pain, and skin infection • Dose Interruptions: 21 (34%) patients – Hemorrhage, increased alanine aminotransferase (ALT), and increased aspartate aminotransferase (AST)

24 Secondary Malignancies

• At the target dose (n=668) in adults – 3 patients developed acute myeloid (AML) – 2 patients developed myelodysplastic syndrome (MDS) (1 subsequently evolved to AML)

• Across the development program (n=822), including pediatric patients – 6 (0.7%) developed secondary malignancies

• Of the 6 patients who developed secondary malignancies: – Time from start of tazemetostat to development of malignancy ranged from 14 months to 4 years (median 27 months) – 5 patients received prior chemotherapy

25 Toxicology Studies

Adult rats • T cell lymphoma with concurrent leukemia led to early death in multiple animals starting 9 weeks into dosing and was present in surviving animals at terminal and recovery necropsy – 28% at the mid dose – 2.5% at the high dose

Juvenile rats • T cell lymphoma led to early deaths in multiple animals starting 11 weeks into dosing and was present in surviving animals at terminal and recovery necropsy 26 Summary of the Major Review Issue

With limited clinical experience and lack of comparative data, FDA is concerned that activity observed in Cohorts 5 and 6 of EZH-202 may not be sufficient to establish the benefit of tazemetostat in patients with epithelioid sarcoma.

Confirmatory study of tazemetostat with doxorubicin compared to doxorubicin alone in patients with epithelioid sarcoma is planned; enrollment has not yet begun. 27 Summary of Data Factor Conclusion

Commonly used Existing therapies for soft tissue sarcoma yield response rates generally <20% with therapies considerable toxicity. Response rate similar for epithelioid sarcoma. Magnitude and Point estimate of ORR with tazemetostat is 11-15%. Limited number of responders duration of effect precludes accurate assessment of DOR. Tumor burden 84% of non-nodal target lesions were 5cm or smaller in longest diameter at baseline. However, this does not completely characterize baseline tumor burden. Mechanism of action Target of tazemetostat (EZH2) may not be particularly relevant for survival in epithelioid sarcoma. Safety 48% of patients with Grade 3-4 adverse event, 37% with serious adverse event. 34% required dose interruption for adverse event; dose reduction or discontinuation was uncommon. Secondary malignancies are an identified risk. Prior experience Tazemetostat not approved for any indication; activity of the drug observed in follicular lymphoma.

28 Discussion and Voting Questions Discussion Question: Discuss whether the evidence from Cohorts 5 and 6 of EZH-202 is sufficient to establish the benefit of tazemetostat in patients with epithelioid sarcoma.

Voting Question: Does the demonstrated benefit of tazemetostat outweigh the risks of the drug in the proposed indication? 29 FDA Review Team

• Meng Li, Clinical Pharmacology Reviewer, OCP/DCPV • Afrouz Nayernama, Acting TL, DPV II, OSE • Michael Tollon, Regulatory Affairs Team leader, ORA • Anamitro Banerjee, Branch Chief, OPQ • Michelle Nadeau-Nguyen, Pharmacovigilance reviewer, DPV II, OSE • Angelica Dorantes, Biopharmaceutics Branch Chief, OPQ • Nam Atiqur Rahman, Director, DCP V, OCP • Ashley Ward, Cross Discipline Team Lead, DO3 • Nan Zheng, Clinical Pharmacology Reviewer, OCP/DPM • Banu Zolnik, Biopharmaceutics Team Leader, OPQ • Olen Stephens, Drug Product Team Leader, OPQ • Barbara Fuller, Patient Labeling Team, DMPP • Peter Waldron, Team Leader, DPV II, OSE • Brad Moriyama, DRISK reviewer, RMA • Quamrul Majumder, Process/Facilities Reviewer, Microbiology, • Caryn McNab, Regulatory Affairs reviewer, ORA OPQ • Chi-Ming (Alice) Tu, Team Leader, DMEPA • Rajeshwari Sridhara, Division Director, OB/DBV • Daniel Jansen, Drug Substance Reviewer, Product Quality, ONDQA • Rakhi Shah, Process/Facilities Team Lead, Microbiology, OPQ • Devi Kozeli, QT/IRT Reviewer, OCP/DPM • Richard (Scott) Swain, Team Leader, OSE/OPE/DEPII • Elizabeth Everhart, DRISK Team Leader, RMA • Roseane Charlab Orbach, GTTG Team Leader, OCP/DPM • Emily Dvorsky, Reviewer, OPDP • Sara Dorff, GTTG Reviewer, OCP/DPM • Fang Tian, Epidemiology Reviewer, OSE/OPE/DEPII • Shamika Brooks, Regulatory Business Process Manager, OPQ • Fengmin Li, Reviewer, CDRH • Shawna Hutchins, Patient Labeling Reviewer, DMPP • Hong Zhao, Clinical Pharmacology Team Leader, OCP/DCPV • Soma Ghosh, Team Leader, CDRH • Janine Stewart, Reviewer, DMEPA • Stacy Shord, Associate Director of Labeling, OOD • Jiang Liu, Pharmacometrics Team Leader, OCP/DPM • Stephanie Aungst, Nonclinical Reviewer, DHOT • Jonathon Vallejo, Statistics Reviewer, OB/DBV • Steven Lemery, Division Director, DO3 • Kaushalkumar Dave, Biopharmaceutics Reviewer, OPQ • Su Tran, Drug Substance Team Lead, Product Quality, ONDQA • Kristin Jarrell, RPM, DO3 • Susan Thompson, Team Leader, OSI • Latonia Ford, Safety Regulatory Project Manager, OSE • Susannah O’Donnell, Team Lead, OPDP • Leslie Doros, Clinical Reviewer, DO3 • Tefsit Bekele, Drug Product Reviewer, OPQ • Lisa Rodriguez, Statistics Team Leader, OB/DBV • Whitney Helms, Nonclinical Team Leader, DHOT • Marc Theoret, Acting Deputy Director, OOD • Yang-Min (Max) Ning, Reviewer, OSI • Melanie Pierce, CPMS, DO3 Chief RPM, ORO, DO3 • Yuan Li Shen, Associate Director, DBV 30 • Yuan Xu, Pharmacometrics Reviewer, OCP/DPM

BACK UP SLIDES SHOWN

32 Interpretation of Stable Disease

• FDA does not consider stable disease in efficacy assessments. – While a response may be attributed directly the activity of a treatment, stable disease may occur with or without treatment.

• In addition, the percentage of patients who experience stable disease by a given time point depends heavily on the natural history of the disease and timing of assessments.

33 Interpretation of Stable Disease

Consider the number of patients who experienced stable disease in the placebo arm of PALETTE1: Pazopanib Placebo N = 246 N = 123 Partial Response2 14 (6%) 0 (0%) Stable disease 164 (67%) 47 (38%) Progression 57 (23%) 70 (57%) Early death 3 (1%) 6 (5%) 1 As reported in Van Der Graaf et. al (2012). Response rate may differ from labeled response rate due to updated data at the time of publishing. 2 Assessed by RECIST v1.0. 34